Biomarkers of regional lung function, coupled with validated low dose methods of assessing anatomic features of the lung are critical to promote discovery and testing of new interventions in COPD and asthma. This proposed bioengineering research partnership seeks to take advantage of the emerging acquisition technique of multi-spectral computed tomography (currently dual energy CT: DECT), careful evaluation of dose lowering methods, and novel approaches to statistical cluster anlaysis to expand the biomarkers used in multi-center studies to identify sub-populations of lung disease. Current CT methods have focused largely on parenchymal destruction, air trapping and airway remodeling. With our recent findings reported in the Proceedings of the National Acedemy of Sciences and the New England Journal of Medicine, there is growing evidence that the etiology of emphysema may be correlated with abnormal vascular responses to inflammation in COPD. To further validate these findings, we focus on multi-spectral CT to simplify the current dynamic CT approach in its assessment of ventilation and perfusion. With DECT we can simplify to a single breath of xenon gas or a slow peripheral injection of iodinated contrast agent to assess regional ventilation or perfused blood volume (PBV). Our approach consists of 5 tightly integrated aims seeking to: 1) establish the minimum dose required to achieve the measurements of importance in defining COPD and asthma sub-populations; 2) use our well characterized CT assessment of pulmonary perfusion and ventilation using dynamic axial imaging to validate metrics from DECT, providing indices of ventilation and perfusion via single breath hold / single lung volume techniques; 3) expand image segmentation of the lung to the pulmonary arterial and venous trees to further link structure to function as well as to reliabily provide a framework for dividing the lung into sublobar segments as the standard region of interest; 4) test the application of a novel statistica approach to cluster analysis such that the measures from quantitative CT fully account for specific phenotypes for disease subgroups and link to a computational fluid dynamics model such that a putative phenotype can be better understood; and finally 5) provide a framework whereby newly developed protocols are harmonized across manufacturers and scanner models, allowing for cross institutional data collection and a means whereby technology is allowed to progress within the context of longitudial studies.